Dynamic threshold determination method and Euclidean clustering method based on point cloud density and spacing

The invention discloses a dynamic threshold determination method based on point cloud density and spacing and a European clustering method, and belongs to the technical field of automatic driving environment perception. The invention aims to solve the problem that the under-segmentation rate of remote obstacle clustering is relatively high due to a fixed threshold value when a uniform distance threshold value is adopted for obstacle recognition in a full range for European clustering. The method comprises the following steps: firstly, establishing a KD-tree according to a laser point cloud after ground segmentation; calculating the average Euclidean distance between the point clouds in each set based on the established KD-tree, and determining a compensation factor e based on the average Euclidean distance of the point cloud set; determining an initial threshold di according to two points in the clustering set; and finally, according to the compensation factor and the initial threshold, determining a distance